Video Game Analysis
Kuldeep Parashar
02/01/2019
0.1 Whats its all About?
This is a exhaustive analytics report for getting a clear insights on the video game industry, from its very primitive stage to the peek of the video game industries.the data set is from Kaggle
This report answers Question on Video Game industries some of them are Stated below
which platform have the most game available??
which company has the most profitable Video game
What share of market is held by companies in different
0.1.1 Introducing Our Source The Data
The data is in Csv(Comma Separated) format,the dimensions are r< dim(df) >. The names of all the column and their meanings are stated below:-
| Atrributes | meanings |
|---|---|
| Rank | Rank of video game |
| Name | Name of video game |
| Platform | Platform for which it is developed |
| Year | Year of release |
| Genre | type of game |
| Publisher | Publisher/Developing Company |
| NA_Sales | North Amrica total Sales |
| EU_Sales | Europe total Sales |
| JP_Sales | Japan Sales |
| Other_Sales | Sales in all other Countries |
| Global_Sales | Global total Sales |
there are missing data in the csv so we have to clean the data and also tidy the data
0.1.2 Data Wrangling
Data Wrangling is the term collectively given to Data Cleaning And Data Tidying in this process do the following things :-
- Check data Consistency,duplicates
- Check for Missing Data
- Check For Outlines
- Found a strong reason before removing Outliers
- Fill the Missing Values
- Fill the the corrupted Data with proper data
- Feature Engineering-process of making new Features
Lets get hands on on to this:-
First converting all the character into factor so that we can easily implement Statistics modelling function and also it would be handy to use them in plotting libraries like ggplot2
now we can see that categorical data are interpreted by R, when we look at the data you will see that ‘N/A’ is used for representing NA, if we did not change it R will not recognize it as a Missing value and we get error prone results.
## Rank Name Platform Year
## Min. : 1 Length:16598 DS :2163 2009 :1431
## 1st Qu.: 4151 Class :character PS2 :2161 2008 :1428
## Median : 8300 Mode :character PS3 :1329 2010 :1259
## Mean : 8301 Wii :1325 2007 :1202
## 3rd Qu.:12450 X360 :1265 2011 :1139
## Max. :16600 PSP :1213 (Other):9868
## (Other):7142 NA's : 271
## Genre Publisher
## Action :3316 Electronic Arts : 1351
## Sports :2346 Activision : 975
## Misc :1739 Namco Bandai Games : 932
## Role-Playing:1488 Ubisoft : 921
## Shooter :1310 Konami Digital Entertainment: 832
## Adventure :1286 (Other) :11529
## (Other) :5113 NA's : 58
## NA_Sales EU_Sales JP_Sales Other_Sales
## Min. : 0.0000 Min. : 0.0000 Min. : 0.00000 Min. : 0.00000
## 1st Qu.: 0.0000 1st Qu.: 0.0000 1st Qu.: 0.00000 1st Qu.: 0.00000
## Median : 0.0800 Median : 0.0200 Median : 0.00000 Median : 0.01000
## Mean : 0.2647 Mean : 0.1467 Mean : 0.07778 Mean : 0.04806
## 3rd Qu.: 0.2400 3rd Qu.: 0.1100 3rd Qu.: 0.04000 3rd Qu.: 0.04000
## Max. :41.4900 Max. :29.0200 Max. :10.22000 Max. :10.57000
##
## Global_Sales
## Min. : 0.0100
## 1st Qu.: 0.0600
## Median : 0.1700
## Mean : 0.5374
## 3rd Qu.: 0.4700
## Max. :82.7400
## Now we will check the consistency of the data, weather tha data inside a column is homogeneous or not, or the data inside column is fisible or not.
taking the mean of the differences between the actualSale calculated by summing up Sales from all countries to the Global_Sale Attributes we get
## [1] 0.0002765393so from here we ca see that the Global_sale atrribute is not correct and has some error init since the value in revenue is in million dollars so there is significant amount which is entered false in the data lets change the value of the Global_sale with the sum of japansale,North America Sale,Europe Sale and others sale
the long tail in the graph clearly states that there are only very few games which have total revenue greater then 75.Most probably these are the most popular game, if not so it may be an outlier.Also we have to check for the duplicacy of the data
## # A tibble: 2,775 x 2
## Name count
## <chr> <int>
## 1 Need for Speed: Most Wanted 12
## 2 FIFA 14 9
## 3 LEGO Marvel Super Heroes 9
## 4 Madden NFL 07 9
## 5 Ratatouille 9
## 6 Angry Birds Star Wars 8
## 7 Cars 8
## 8 FIFA 15 8
## 9 FIFA Soccer 13 8
## 10 Lego Batman 3: Beyond Gotham 8
## # ... with 2,765 more rowsso here we can see that there are 2,775 videogames which are being published more then once, surely these game must have great revenue thats why there are multiple release
in the next section we will analysis the trend and try to find the correlations and give ans to various Curious Questions too.
0.1.3 Univariate
So here we see that how the data is being spread and its central tendencies to get an direct insights of the data
0.1.3.1 Yearly Increase in Videogame Development
## 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994
## 9 46 36 17 14 14 21 16 15 17 16 41 43 60 121
## 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009
## 219 263 289 379 338 349 482 829 775 744 936 1008 1201 1428 1431
## 2010 2011 2012 2013 2014 2015 2016 2017 2020
## 1257 1136 655 546 580 614 342 3 10.1.3.2 Genre wise No. of Game Developed
This graphs shows which, which genre have most no. games in it .Action Games are at the tops followed by Sports, here interesting insights is that, miscellenious game are 3rd highest ranking.
This graphs shows which, which genre have most no. games in it .Action Games are at the tops followed by Sports, here interesting insights is that, miscellenious game are 3rd highest ranking.0.1.3.3 game Genre distribution on Countries
lets see distribution of companies developing game in a Genre.
0.1.3.4 Sales Country wise Analysis
here we can clearly say that most the sales come from the North America but if we think s point of view of Marketing its not a great metric, as we know that japan has less population so if we incorporate that factor into metric things may be different
0.1.4 Bivariate
Here we will the effects of on variable as the second variable changes an
0.2 Get Some insights
0.2.1 Top 10 Revenue generating Games
todo the leaft work and to mark things to geates and to let things
| Name | total_sale |
|---|---|
| Wii Sports | 82.74 |
| Grand Theft Auto V | 55.92 |
| Super Mario Bros. | 45.31 |
| Tetris | 35.84 |
| Mario Kart Wii | 35.82 |
| Wii Sports Resort | 33.00 |
| Pokemon Red/Pokemon Blue | 31.37 |
| Call of Duty: Modern Warfare 3 | 30.83 |
| New Super Mario Bros. | 30.01 |
| Call of Duty: Black Ops II | 29.72 |
So till 2016 these games have the most global revenue, Wii sports which is nentando game is on the top and google search link also state this (Quite intresting actually) GrandTheft Auto is on 2 followed by Super Mario
0.2.2 Top 5 Revenue Generating Genres
For a Game developer finding the sweet spot is important to make revenue in such a competative market, lets first find which genre generate maximum revenue and after that we will find that which genre has least compettion that is total revenue devided by total no of video game companies making games on that genre| Genre | total_revenue |
|---|---|
| Action | 1722.84 |
| Sports | 1309.24 |
| Shooter | 1026.20 |
| Role-Playing | 923.83 |
| Platform | 829.13 |
| Misc | 789.87 |
| Racing | 726.76 |
| Fighting | 444.05 |
| Simulation | 389.98 |
| Puzzle | 242.21 |
| Adventure | 234.59 |
| Strategy | 173.27 |
here we can see that Misc genre is on 3 position in no of count of games but in case revenue genration its far from Action and Sports and all.
0.2.3 Sweet Spot!!
now lets find that which genre has the least no of videogame devloped init
metric would be like to make total revenue generate by a genre devid eby the total no a video game in the genre
| Genre | total_revenue | count | ease_metric |
|---|---|---|---|
| Platform | 829.13 | 875 | 0.9475771 |
| Shooter | 1026.20 | 1282 | 0.8004680 |
| Role-Playing | 923.83 | 1470 | 0.6284558 |
| Racing | 726.76 | 1225 | 0.5932735 |
| Sports | 1309.24 | 2304 | 0.5682465 |
| Fighting | 444.05 | 836 | 0.5311603 |
| Action | 1722.84 | 3251 | 0.5299416 |
| Misc | 789.87 | 1686 | 0.4684875 |
| Simulation | 389.98 | 848 | 0.4598821 |
| Puzzle | 242.21 | 570 | 0.4249298 |
| Strategy | 173.27 | 670 | 0.2586119 |
| Adventure | 234.59 | 1274 | 0.1841366 |
some intresting facts again as we can see that platorm and Shotter game are genrating great average revenue per game, action game clearly get out of top 5 and its clearly show that there are alot number of games,